CN103020845A - Mobile application pushing method and system - Google Patents

Abstract

The invention provides a mobile application pushing method. The mobile application pushing method includes determining more than one mobile application category with the highest relevancy to mobile applications designed by a user according to relevancy of pre-generated mobile application categories; calculating recommendation levels of the mobile applications under the mobile application categories according to pre-generated weighted values of the mobile applications; extracting the mobile application with higher rankings of the recommendation levels under each mobile application category; and selecting more than one mobile application which has the highest recommendation level as a recommendation result according to the number of pre-set recommendation results, and pushing the more than one mobile application to users. The invention further provides a mobile application pushing system. According to the mobile application pushing method and the system, diversity of the recommended mobile applications can be effectively improved.

Description

A kind of method for pushing and system that moves application

[technical field]

The present invention relates to the internet, applications field, relate in particular to a kind of method for pushing and system that moves application.

[background technology]

At present, mobile application shop all can push some to the user and move application when user's download or browse application, in order to recommend mobile the application to the user, method for pushing is that the User history log counts the correlativity between mobile the application, then according to correlativity and utilize the proposed algorithms such as neighbour, collaborative filtering to produce recommendation results, so prior art all is as recommending according to recommending mobile the application with the correlativity between mobile the application.

Therefore, there is following problem in the mobile way of recommendation of using at present:

1, since all be according to the correlativity between mobile the application as recommending foundation so that the mobile content of using of recommending is too similar, can't recommend diversified mobile the application to the user, thereby can't excite the user to the mobile demand of using.

2, because newly-increased mobile the application do not have user's history log, therefore can't add up newly-increased mobile use and other move correlativity between the application, therefore can't be when the user checks or downloads mobile the application, newly-increased mobile the application recommended the user, can't solve the newly-increased mobile cold start-up problem of using.

[summary of the invention]

The invention provides a kind of method for pushing and system that moves application, the mobile diversity of using of can Effective Raise recommending.

Concrete technical scheme of the present invention is as follows:

According to one preferred embodiment of the present invention, a kind of method for pushing that moves application comprises:

According to the degree of correlation of the mobile applicating category that generates in advance, determine one or more the mobile applicating category the highest with the mobile mobile applicating category degree of correlation of using of user's appointment;

According to the mobile weighted value of using that generates in advance, calculate the mobile recommendation degree of using under the described mobile applicating category;

Extract each and move forward mobile application of recommendation degree rank under the applicating category, the mobile application more than that the recommendation degree is the highest in mobile application that the default recommendation results number of foundation will be extracted is pushed to the user as recommendation results.

In the said method, the method that generates in advance the degree of correlation of mobile applicating category is:

Obtain the mobile classification information of using according to the mobile ontology library of using, according to the mobile classification information of using mobile application that the user checks, downloads and uses classified;

According to the user that obtains mobile use check or download in the shop mobile information, the user who uses uses the mobile duration information of using and mobile application that generate in advance between the degree of correlation, calculate the degree of correlation between the mobile applicating category.

In the said method, the method that generates in advance the degree of correlation between mobile the application is:

In mobile application shop, check or download the information and the user that move application according to described user and use the mobile duration information of using, and move the degree of correlation between the application in the mobile set of applications of utilizing following formula calculating user to check, download and use:

$R({\mathrm{app}}_m,{\mathrm{app}}_n)=\underset{u=1}{\overset{U}{\mathrm{\Σ}}}{w}_u\×\frac{s_{{\mathrm{app}}_m}\×(k_1+1)}{s_{{\mathrm{app}}_m}+K}\×\frac{s_{{\mathrm{app}}_n}\×(k_2+1)}{s_{{\mathrm{app}}_n}+k_2}$

Wherein, R (app _m, app _n) represent to move in the mobile set of applications and use app _mWith the mobile app that uses _nBetween the degree of correlation, U represents to use simultaneously the mobile app of application _mWith the mobile app that uses _nUser set,

With Represent that respectively the user gathers that user u is app among the U _mAnd app _nThe score value that distributes; w _uThe expression user gathers the weight of user u among the U,

k ₁Equal 2, k ₂Equal 1.2, b and equal 0.75, n _uThe expression user gathers the mobile sum of using that user u checks, downloads and uses among the U, n _AvgThe mobile sum of using of the mobile sum of using that expression user u checks, download and the mean value that uses the mobile sum of using.

In the said method, described user gathers the weight w of user u among the U _uFor Wherein, N represents the mobile sum of using in the mobile set of applications, n _uThe expression user gathers the mobile sum of using that user u checks, downloads and uses among the U.

In the said method, described user is the mobile app that uses _mThe score value that distributes

For $s_{{\mathrm{app}}_m}=s_1\×{\mathrm{read}}_{{\mathrm{app}}_m}+s_2\×{\mathrm{download}}_{{\mathrm{app}}_m}+s_3\×{\mathrm{usetime}}_{{\mathrm{app}}_m};$

Wherein, s ₁Equal 1, s ₂Equal 2, s ₃Equal 1; When the user checks the mobile app of application _mThe time Equal 1, when the user does not check the mobile app of application _mThe time

Equal 0; When the user downloads the mobile app of application _mThe time Equal 1, when the user does not download the mobile app of application _mThe time

Equal 0;

Equal the user and use the mobile app of application _mDuration.

In the said method, the method for the degree of correlation between the mobile applicating category of described calculating is:

Utilize following formula to calculate mobile applicating category concept _iWith mobile applicating category concept _jBetween the degree of correlation:

$R({\mathrm{concept}}_i,{\mathrm{concept}}_j)=\underset{{\mathrm{concept}}_{{\mathrm{app}}_n}={\mathrm{concept}}_j}{\overset{U}{\underset{{\mathrm{concept}}_{{\mathrm{app}}_m}={\mathrm{concept}}_i,}{\mathrm{\Σ}}}}\frac{f_{{\mathrm{app}}_m{\mathrm{app}}_n}}{f_{{\mathrm{app}}_m}+f_{{\mathrm{app}}_n}}\×R({\mathrm{app}}_m,{\mathrm{app}}_n)$

Wherein, concept _iAnd concept _jBe respectively the mobile app of application _mWith the mobile app that uses _nAffiliated mobile applicating category, R (app _m, app _n) be the mobile app of application in the mobile set of applications _mWith the mobile app that uses _nBetween the degree of correlation,

The mobile app of application is checked in expression _mThe user, download the mobile app of application _mThe user and use the mobile app of application _mUser's sum,

The mobile app of application is checked in expression _nThe user, download the mobile app of application _nThe user and use the mobile app of application _nUser's sum, The mobile app of application is checked in expression _mThe user, download the mobile app of application _mThe user and use the mobile app of application _mThe user set with check the mobile app of application _nThe user, download the mobile app of application _nThe user or use the mobile app of application _nUser's intersection of sets total number of users that collection comprises.

In the said method, the method that generates in advance the mobile weighted value of using is:

Utilize following formula to calculate the mobile app of application _mAt mobile applicating category concept _iIn weighted value:

$w_{{\mathrm{concept}}_i{\mathrm{app}}_m}=k_1\×\frac{r_{{\mathrm{app}}_m}}{r_{{\mathrm{concept}}_i}}+k_2\×\frac{d_{{\mathrm{app}}_m}}{d_{{\mathrm{concept}}_i}}+k_3\×\frac{u_{{\mathrm{app}}_m}}{u_{{\mathrm{concept}}_i}}$

Wherein,

Be respectively the mobile app of application _mIn user's history log by the total degree checked, the total degree that is downloaded, total duration of being used;

Be respectively mobile applicating category concept _iTotal duration that lower all movements are applied in the total degree of being checked in user's history log, the total degree that is downloaded, are used; k ₁Equal 0.2, k ₂Equal 0.4, k ₃Equal 0.4.

In the said method, the method also comprises:

With the mobile ontology library that adds mobile application to of using that increases newly in the mobile application shop, and be newly-increased mobile classification information and the attribute information that marks correspondence of using;

On duty with default decay factor with the mobile average weight of using that rank under the mobile applicating category under newly-increased mobile the application is forward, the mobile weighted value of using that obtains increasing newly.

In the said method, the method for the mobile recommendation degree of using is under the described mobile applicating category of described calculating:

Utilize following formula to calculate that each moves the recommendation degree of application in the mobile applicating category:

${\mathrm{rec}}_{{\mathrm{app}}_m{\mathrm{app}}_n}=R({\mathrm{concept}}_i,{\mathrm{concept}}_j)\×w_{{\mathrm{concept}}_j{\mathrm{app}}_n}+k\×\mathrm{comatt}({\mathrm{app}}_m,{\mathrm{app}}_n)$

Wherein,

For specifying the mobile app of application _mThe time recommend the mobile app of application to the user _nThe recommendation degree, the mobile app that uses _mAffiliated mobile applicating category is concept _i, the mobile app that uses _nAffiliated mobile applicating category is concept _j, mobile applicating category concept _jBe positioned at and mobile applicating category concept _iIn the highest mobile applicating category of the degree of correlation, R (concept _i, concept _j) be mobile applicating category concept _iWith mobile applicating category concept _jBetween the degree of correlation,

Be mobile applicating category concept _jThe lower mobile app that uses _nWeighted value, comatt (app _m, app _n) use app for moving _nWith the mobile app that uses _mThe number of same alike result, k equals 2.

A kind of supplying system that moves application comprises: statistic unit, the first computing unit, push unit; Wherein,

Statistic unit is used for the degree of correlation according to the mobile applicating category that generates in advance, determines one or more the mobile applicating category the highest with the mobile mobile applicating category degree of correlation of using of user's appointment;

The first computing unit is used for calculating the mobile recommendation degree of using under the described mobile applicating category according to the mobile weighted value of using that generates in advance;

Push unit is used for extracting each and moves forward mobile application of recommendation degree rank under the applicating category, and the mobile application more than that the recommendation degree is the highest in mobile application that the default recommendation results number of foundation will be extracted is pushed to the user as recommendation results.

In the said system, this system also comprises: the second computing unit that is used for generating in advance the degree of correlation of mobile applicating category:

The degree of correlation that the second computing unit generates mobile applicating category in advance specifically comprises: obtain the mobile classification information of using according to the mobile ontology library of using, according to the mobile classification information of using mobile application that the user checks, downloads and uses classified; According to the user that obtains mobile use check or download in the shop mobile information, the user who uses uses the mobile duration information of using and mobile application that generate in advance between the degree of correlation, calculate the degree of correlation between the mobile applicating category.

In the said system, this system also comprises for the 3rd computing unit that generates in advance the degree of correlation between mobile the application;

The degree of correlation that described the 3rd computing unit generates between mobile the application in advance specifically comprises:

In mobile application shop, check or download the information and the user that move application according to described user and use the mobile duration information of using, and move the degree of correlation between the application in the mobile set of applications of utilizing following formula calculating user to check, download and use:

$R({\mathrm{app}}_m,{\mathrm{app}}_n)=\underset{u=1}{\overset{U}{\mathrm{\Σ}}}{w}_u\×\frac{s_{{\mathrm{app}}_m}\×(k_1+1)}{s_{{\mathrm{app}}_m}+K}\×\frac{s_{{\mathrm{app}}_n}\×(k_2+1)}{s_{{\mathrm{app}}_n}+k_2}$

Wherein,

Represent the mobile app of application in the mobile set of applications _mWith the mobile app that uses _nBetween the degree of correlation, U represents to use simultaneously the mobile app of application _mWith the mobile app that uses _nUser set,

With

Represent that respectively the user gathers that user u is app among the U _mAnd app _nThe score value that distributes; w _uThe expression user gathers the weight of user u among the U,

k ₁Equal 2, k ₂Equal 1.2, b and equal 0.75, n _uThe expression user gathers the mobile sum of using that user u checks, downloads and uses among the U, n _AvgThe mobile sum of using of the mobile sum of using that expression user u checks, download and the mean value that uses the mobile sum of using.

In the said system, described user gathers the weight w of user u among the U _uFor

Wherein, N represents the mobile sum of using in the mobile set of applications, n _uThe expression user gathers the mobile sum of using that user u checks, downloads and uses among the U.

In the said system, described user is the mobile app that uses _mThe score value that distributes

For $s_{{\mathrm{app}}_m}=s_1\×{\mathrm{read}}_{{\mathrm{app}}_m}+s_2\×{\mathrm{download}}_{{\mathrm{app}}_m}+s_3\×{\mathrm{usetime}}_{{\mathrm{app}}_m};$

Wherein, s ₁Equal 1, s ₂Equal 2, s ₃Equal 1; When the user checks the mobile app of application _mThe time

Equal 1, when the user does not check the mobile app of application _mThe time

Equal 0; When the user downloads the mobile app of application _mThe time

Equal 1, when the user does not download the mobile app of application _mThe time

Equal 0;

Equal the user and use the mobile app of application _mDuration.

In the said system, the degree of correlation that described the second computing unit calculates between the mobile applicating category specifically comprises:

Utilize following formula to calculate mobile applicating category concept _iWith mobile applicating category concept _jBetween the degree of correlation:

$R({\mathrm{concept}}_i,{\mathrm{concept}}_j)=\underset{{\mathrm{concept}}_{{\mathrm{app}}_n}={\mathrm{concept}}_j}{\overset{U}{\underset{{\mathrm{concept}}_{{\mathrm{app}}_m}={\mathrm{concept}}_i,}{\mathrm{\Σ}}}}\frac{f_{{\mathrm{app}}_m{\mathrm{app}}_n}}{f_{{\mathrm{app}}_m}+f_{{\mathrm{app}}_n}}\×R({\mathrm{app}}_m,{\mathrm{app}}_n)$

Wherein, concept _iAnd concept _jBe respectively the mobile app of application _mWith the mobile app that uses _nAffiliated mobile applicating category, R (app _m, app _n) be the mobile app of application in the mobile set of applications _mWith the mobile app that uses _nBetween the degree of correlation,

The mobile app of application is checked in expression _mThe user, download the mobile app of application _mThe user and use the mobile app of application _mUser's sum,

The mobile app of application is checked in expression _nThe user, download the mobile app of application _nThe user and use the mobile app of application _nUser's sum,

The mobile app of application is checked in expression _mThe user, download the mobile app of application _mThe user and use the mobile app of application _mThe user set with check the mobile app of application _nThe user, download the mobile app of application _nThe user or use the mobile app of application _nUser's intersection of sets total number of users that collection comprises.

In the said system, this system also comprises for the 4th computing unit that generates in advance the mobile weighted value of using;

Described the 4th computing unit generates in advance the mobile weighted value of using and specifically comprises:

Utilize following formula to calculate the mobile app of application _mAt mobile applicating category concept _iIn weighted value:

$w_{{\mathrm{concept}}_i{\mathrm{app}}_m}=k_1\×\frac{r_{{\mathrm{app}}_m}}{r_{{\mathrm{concept}}_i}}+k_2\×\frac{d_{{\mathrm{app}}_m}}{d_{{\mathrm{concept}}_i}}+k_3\×\frac{u_{{\mathrm{app}}_m}}{u_{{\mathrm{concept}}_i}}$

Wherein,

Be respectively the mobile app of application _mIn user's history log by the total degree checked, the total degree that is downloaded, total duration of being used; Be respectively mobile applicating category concept _iTotal duration that lower all movements are applied in the total degree of being checked in user's history log, the total degree that is downloaded, are used; k ₁Equal 0.2, k ₂Equal 0.4, k ₃Equal 0.4.

In the said system, this system also comprises: updating block;

Described updating block adds the ontology library of mobile application to for the mobile application that mobile application shop is increased newly, and is that classification information and the attribute information that marks correspondence used in newly-increased moving;

Described the 4th computing unit, also on duty with default decay factor with the forward mobile average weight of using of rank under the affiliated mobile applicating category of newly-increased mobile application, the weighted value that moves application that obtains increasing newly.

In the said system, the recommendation degree that described the first computing unit calculates mobile application under the described mobile applicating category specifically comprises:

Utilize following formula to calculate that each moves the recommendation degree of application in the mobile applicating category:

${\mathrm{rec}}_{{\mathrm{app}}_m{\mathrm{app}}_n}=R({\mathrm{concept}}_i,{\mathrm{concept}}_j)\×w_{{\mathrm{concept}}_j{\mathrm{app}}_n}+k\×\mathrm{comatt}({\mathrm{app}}_m,{\mathrm{app}}_n)$

Wherein,

For specifying the mobile app of application _mThe time recommend the mobile app of application to the user _nThe recommendation degree, the mobile app that uses _mAffiliated mobile applicating category is concept _i, the mobile app that uses _nAffiliated mobile applicating category is concept _j, mobile applicating category concept _jBe positioned at and mobile applicating category concept _iIn the highest mobile applicating category of the degree of correlation, R (concept _i, concept _j) be mobile applicating category concept _iWith mobile applicating category concept _jBetween the degree of correlation,

Be mobile applicating category concept _jThe lower mobile app that uses _nWeighted value, comatt (app _m, app _n) use app for moving _nWith the mobile app that uses _mThe number of same alike result, k equals 2.

As can be seen from the above technical solutions, technical scheme provided by the invention has following beneficial effect:

Extract forward mobile application of recommendation degree rank under the higher mobile applicating category of the degree of correlation, and mobile application that wherein degree of recommendation is the highest recommended the user, so, guaranteed the diversity of the mobile applicating category recommended, thus the mobile diversity of using of can Effective Raise recommending.

[description of drawings]

Fig. 1 is the schematic flow sheet that the present invention realizes the preferred embodiment of the mobile method for pushing of using;

Fig. 2 is the structural representation that the present invention realizes the preferred embodiment of the mobile supplying system of using.

[embodiment]

Basic thought of the present invention is: according to the degree of correlation of the mobile applicating category that generates in advance, determine one or more the mobile applicating category the highest with the mobile mobile applicating category degree of correlation of using of user's appointment; According to the mobile weighted value of using that generates in advance, calculate the mobile recommendation degree of using under the described mobile applicating category; Extract each and move forward mobile application of recommendation degree rank under the applicating category, the mobile application more than that the recommendation degree is the highest in mobile application that the default recommendation results number of foundation will be extracted is pushed to the user as recommendation results.

In order to make the purpose, technical solutions and advantages of the present invention clearer, describe the present invention below in conjunction with the drawings and specific embodiments.

The invention provides a kind of method for pushing that moves application, Fig. 1 is the schematic flow sheet that the present invention realizes the preferred embodiment of the mobile method for pushing of using, and as shown in Figure 1, the preferred embodiment may further comprise the steps:

Step 101, User checks or download the mobile information of using in mobile application shop and the user uses the mobile duration information of using, the degree of correlation in the mobile set of applications that the calculating user checks, downloads and uses between mobile the application.

Concrete, the mobile data platform of using the shop can be stored the user and use the user's history log that moves when using the shop, data platform is with text formatting storage user history log, and per hour the text of the user's history log in hour is saved in the same file as unit; Described user's history log comprises that the user checks or download the mobile information of using in mobile application shop and the user uses the mobile duration information of using; Wherein, the user checks or downloads the mobile information of using and comprise that user ID (UID), this user use the mobile sign (packageID) of using and this user that check or download in the shop and use the mobile time of using of checking or downloading in the shop mobile mobile mobile the application in the shop; Described user uses the mobile duration information of using to comprise user ID (UID).Mobile sign (packageID) and this user who uses that this user uses uses this to move the duration of application.

In mobile application shop, check or download the information and the user that move application according to described user and use the mobile duration information of using, and move the degree of correlation between the application in the mobile set of applications of utilizing formula (1) calculating user to check, download and use:

$R({\mathrm{app}}_m,{\mathrm{app}}_n)=\underset{u=1}{\overset{U}{\mathrm{\Σ}}}{w}_u\×\frac{s_{{\mathrm{app}}_m}\×(k_1+1)}{s_{{\mathrm{app}}_m}+K}\×\frac{s_{{\mathrm{app}}_n}\×(k_2+1)}{s_{{\mathrm{app}}_n}+k_2}---\left(1\right)$

Wherein, R (app _m, app _n) represent to move in the mobile set of applications and use app _mWith the mobile app that uses _nBetween the degree of correlation, U represents to use simultaneously the mobile app of application _mWith the mobile app that uses _nUser set,

With

Represent that respectively the user gathers that user u is app among the U _mAnd app _nThe score value that distributes; w _uThe expression user gathers the weight of user u among the U, can utilize formula (2) to calculate w _u:

$w_u=\log \frac{N-n_u+0.5}{n_u+0.5}---\left(2\right)$

Wherein, N represents the mobile sum of using in the mobile set of applications, n _uThe expression user gathers the mobile mobile sum of using of using and using that mobile that user u checks among the U used, downloads.

Described

Or

Utilize formula (3) to obtain:

$s_{{\mathrm{app}}_m}=s_1\×{\mathrm{read}}_{{\mathrm{app}}_m}+s_2\×{\mathrm{download}}_{{\mathrm{app}}_m}+s_3\×{\mathrm{usetime}}_{{\mathrm{app}}_m}---\left(3\right)$

Wherein, s ₁, s ₂And s ₃Represent respectively the mobile mobile mobile basic mark of using of using and using of using, downloading that user u checks, in this preferred embodiment, s ₁Equal 1, s ₂Equal 2, s ₃Equal 1;

Whether expression user u checks the mobile app of application _m, if so, Equal 1, if not,

Equal 0;

Whether expression user u downloads the mobile app of application _m, if so,

Equal 1, if not,

Equal 0;

The expression user uses the mobile app of application _mDuration, here, the user uses the mobile app of application _mDuration take minute as unit;

Wherein, k ₁And k ₂Be the adjustment factor, in this preferred embodiment, k ₁Equal 2, k ₂Equal 1.2, K and utilize formula

Obtain; Wherein, b is for adjusting the factor, and in this preferred embodiment, b equals 0.75, n _AvgThe mobile sum of using of the mobile sum of using that expression user u checks, download and the mean value that uses the mobile sum of using.

In this preferred embodiment, the degree of correlation between the application is moved in calculating that can the cycle, for example, can extract in morning every day user's history log in previous month, calculates the degree of correlation between the mobile application according to this user's history log.

Step 102 obtains the mobile classification information of using in the mobile set of applications according to the mobile ontology library of using, and according to the mobile classification information of using mobile the application is classified; User mobile use check or download in the shop mobile information, the user who uses uses the mobile duration information of using and mobile application that calculate between relatedness computation move the degree of correlation between the applicating category.

Concrete, the mobile ontology library of using is take the mobile sign (packageID) of using as unit, and comprises title, classification information and the attribute information that this moves the sign correspondence of application, for example, the mobile ontology library of using can be as shown in table 1:

Table 1

After the degree of correlation in having calculated mobile set of applications between mobile the application, according to mobile being identified in the mobile ontology library of using of using in the mobile set of applications, obtain the classification information that each moves application, then according to the mobile classification information of using mobile the application classified, obtain mobile applicating category more than corresponding to mobile set of applications, according to the degree of correlation between mobile application that calculates, and utilize formula (4) to calculate the degree of correlation between the mobile applicating category:

$R({\mathrm{concept}}_i,{\mathrm{concept}}_j)=\underset{{\mathrm{concept}}_{{\mathrm{app}}_n}={\mathrm{concept}}_j}{\overset{U}{\underset{{\mathrm{concept}}_{{\mathrm{app}}_m}={\mathrm{concept}}_i,}{\mathrm{\Σ}}}}\frac{f_{{\mathrm{app}}_m{\mathrm{app}}_n}}{f_{{\mathrm{app}}_m}+f_{{\mathrm{app}}_n}}\×R({\mathrm{app}}_m,{\mathrm{app}}_n)---\left(4\right)$

Wherein, concept _iAnd concept _jExpression is moved and is used app respectively _mWith the mobile app that uses _nAffiliated mobile applicating category, R (app _m, app _n) the mobile app that uses in the mobile set of applications that calculates of representation formula (1) _mWith the mobile app that uses _nBetween the degree of correlation, R (concept _i, concept _j) the mobile applicating category concept of expression _iWith mobile applicating category concept _jBetween the degree of correlation;

The mobile app of application is checked in expression _mThe user, download the mobile app of application _mThe user and use the mobile app of application _mUser's sum,

The mobile app of application is checked in expression _nThe user, download the mobile app of application _nThe user and use the mobile app of application _nUser's sum; Here, can User history log statistics check the mobile user who uses, download and move the user who uses and the sum that uses the mobile user who uses;

The mobile app of application is checked in expression _mThe user, download the mobile app of application _mThe user and use the mobile app of application _mThe user set with check the mobile app of application _nThe user, download the mobile app of application _nThe user or use the mobile app of application _nUser's intersection of sets total number of users that collection comprises.

Step 103 with the mobile ontology library that adds mobile application to of using that increases newly in the mobile application shop, and is newly-increased mobile classification information and the attribute information that marks correspondence of using.

Concrete, after having calculated the degree of correlation of mobile applicating category, the mobile application that increases newly in the mobile application shop can also be added in the ontology library, for newly-increased mobile application distributes packageID, and corresponding classification information and the attribute information of mark; Wherein, can utilize mobile mobile title and the brief introduction of using that automatic marking system provides according to the mobile owner of application of using, be newly-increased mobile automatic marking classification information and the attribute information used.

Step 104, User checks or download the mobile information of using in mobile application shop and the user uses the mobile duration information of using, the non-newly-increased mobile weighted value of using under the mobile applicating category in the calculating ontology library; On duty with default decay factor with the mobile average weight of using that rank under the mobile applicating category under newly-increased mobile the application is forward, the mobile weighted value of using that obtains increasing newly.

Concrete, utilize formula (5) to calculate and move the weighted value of using in the mobile ontology library of using under the mobile applicating category:

$w_{{\mathrm{concept}}_i{\mathrm{app}}_m}=k_1\×\frac{r_{{\mathrm{app}}_m}}{r_{{\mathrm{concept}}_i}}+k_2\×\frac{d_{{\mathrm{app}}_m}}{d_{{\mathrm{concept}}_i}}+k_3\×\frac{u_{{\mathrm{app}}_m}}{u_{{\mathrm{concept}}_i}}---\left(5\right)$

Wherein,

The mobile app that uses of expression _mAt mobile applicating category concept _iIn weighted value,

The mobile app that uses of expression _mThe total degree of in user's history log, being checked,

The mobile app that uses of expression _mThe total degree that in user's history log, is downloaded,

The mobile app that uses of expression _mThe total duration that in user's history log, is used,

Unit be minute;

Represent mobile applicating category concept _iLower all movements are applied in the total degree of being checked in user's history log;

Represent mobile applicating category concept _iLower all movements are applied in the total degree that is downloaded in user's history log,

Represent mobile applicating category concept _iLower all movements are applied in the total duration that is used in user's history log; k ₁The mobile app that uses of expression _mIn user's history log, checked corresponding factor of influence, k ₂The mobile app that uses of expression _mIn user's history log, be downloaded corresponding factor of influence, k ₃The mobile app that uses of expression _mThe factor of influence of total duration in user's history log, in this preferred embodiment, k ₁Equal 0.2, k ₂Equal 0.4, k ₃Equal 0.4; The sign that storing mobile is used and the corresponding relation that moves the weighted value of using.

Wherein, if the mobile app that uses _mBe mobile application newly-increased in the mobile ontology library of using, then newly-increased mobile the application adopted the weighted value of giving tacit consent to, and the computing method of the weighted value of acquiescence are for utilizing mobile applicating category concept _iThe mean value of three mobile weighted values of using that middle weighted value is the highest multiply by a decay factor, and in this preferred embodiment, this decay factor equals 0.4.

Step 105 when receiving mobile application of user's appointment, is determined one or more the mobile applicating category the highest with the mobile mobile applicating category degree of correlation of using of appointment according to the degree of correlation of mobile applicating category; According to the mobile weighted value of using, calculate the mobile recommendation degree of using under the described mobile applicating category; Extract each and move forward mobile application of recommendation degree rank under the applicating category, the mobile application more than that the recommendation degree is the highest in mobile application that the default recommendation results number of foundation will be extracted is pushed to the user as recommendation results.

Concrete, when the user searches for or download the mobile app of application mobile the application in the shop _mThe time, inquire about according to mobile being identified in the mobile ontology library of using of using of this appointment, obtain the mobile app of application _mMobile applicating category concept _i, foundation and mobile applicating category concept _iThe descending order of the degree of correlation the mobile applicating category in the mobile ontology library of using is sorted, then according to default recommendation results number n, extract the most forward 2n of a degree of correlation rank mobile applicating category.

Each moves the recommendation degree of application in 2n the mobile applicating category that utilizes formula (6) calculating to extract:

${\mathrm{rec}}_{{\mathrm{app}}_m{\mathrm{app}}_n}=R({\mathrm{concept}}_i,{\mathrm{concept}}_j)\×w_{{\mathrm{concept}}_j{\mathrm{app}}_n}+k\×\mathrm{comatt}({\mathrm{app}}_m,{\mathrm{app}}_n)---\left(6\right)$

Wherein,

For specifying the mobile app of application _mThe time recommend the mobile app of application to the user _nThe recommendation degree, the mobile app that uses _mAffiliated mobile applicating category is concept _i, the mobile app that uses _nAffiliated mobile applicating category is concept _j, mobile applicating category concept _jBe positioned at and mobile applicating category concept _iIn the individual mobile applicating category of the 2n that the degree of correlation is the highest, R (concept _i, concept _j) be mobile applicating category concept _iWith mobile applicating category concept _jBetween the degree of correlation, Be mobile applicating category concept _jThe lower mobile app that uses _nWeighted value, comatt (app _m, app _n) use app for moving _nWith the mobile app that uses _mThe number of same alike result, k is factor of influence, in this preferred embodiment, k equals 2.

After calculating the mobile recommendation degree of using, the User history log, the mobile application that calculates the recommendation degree screened, deleting wherein, the user downloaded or used mobile the application, take mobile applicating category as unit, the order descending according to the recommendation degree sorts to mobile application that each moves under the applicating category, then extract the mobile application that each moves recommendation degree rank front two under the applicating category, sort to move the mobile criteria in application recommendation degree order from high to low that applicating category extracts from each, according to default recommendation results number n, the mobile recommendation results of using as mobile of using with coming front n is pushed to the user with this recommendation results.

For realizing said method, the present invention also provides a kind of supplying system that moves application, Fig. 2 is the structural representation that the present invention realizes the preferred embodiment of the mobile supplying system of using, and as shown in Figure 2, this system comprises: statistic unit 20, the first computing unit 21, push unit 22; Wherein,

Statistic unit 20 is used for the degree of correlation according to the mobile applicating category that generates in advance, determines one or more the mobile applicating category the highest with the mobile mobile applicating category degree of correlation of using of user's appointment;

The first computing unit 21 is used for calculating the mobile recommendation degree of using under the described mobile applicating category according to the mobile weighted value of using that generates in advance;

Push unit 22 is used for extracting each and moves forward mobile application of recommendation degree rank under the applicating category, and the mobile application more than that the recommendation degree is the highest in mobile application that the default recommendation results number of foundation will be extracted is pushed to the user as recommendation results.

This system also comprises: the second computing unit 23 that is used for generating in advance the degree of correlation of mobile applicating category;

The degree of correlation that the second computing unit 23 generates mobile applicating category in advance specifically comprises: obtain the mobile classification information of using according to the mobile ontology library of using, according to the mobile classification information of using mobile application that the user checks, downloads and uses classified; According to the user that obtains mobile use check or download in the shop mobile information, the user who uses uses the mobile duration information of using and mobile application that generate in advance between the degree of correlation, calculate the degree of correlation between the mobile applicating category.

This system also comprises for the 3rd computing unit 24 that generates in advance the degree of correlation between mobile the application;

The degree of correlation that described the 3rd computing unit 24 generates between mobile the application in advance specifically comprises:

In mobile application shop, check or download the information and the user that move application according to described user and use the mobile duration information of using, and move the degree of correlation between the application in the mobile set of applications of utilizing following formula calculating user to check, download and use:

$R({\mathrm{app}}_m,{\mathrm{app}}_n)=\underset{u=1}{\overset{U}{\mathrm{\Σ}}}{w}_u\×\frac{s_{{\mathrm{app}}_m}\×(k_1+1)}{s_{{\mathrm{app}}_m}+K}\×\frac{s_{{\mathrm{app}}_n}\×(k_2+1)}{s_{{\mathrm{app}}_n}+k_2}$

Wherein, R (app _m, app _n) represent to move in the mobile set of applications and use app _mWith the mobile app that uses _nBetween the degree of correlation, U represents to use simultaneously the mobile app of application _mWith the mobile app that uses _nUser set,

With

Represent that respectively the user gathers that user u is app among the U _mAnd app _nThe score value that distributes; w _uThe expression user gathers the weight of user u among the U,

k ₁Equal 2, k ₂Equal 1.2, b and equal 0.75, n _uThe expression user gathers the mobile sum of using that user u checks, downloads and uses among the U, n _AvgThe mobile sum of using of the mobile sum of using that expression user u checks, download and the mean value that uses the mobile sum of using.

Wherein, described user gathers the weight w of user u among the U _uFor

Wherein, N represents the mobile sum of using in the mobile set of applications, n _uThe expression user gathers the mobile sum of using that user u checks, downloads and uses among the U.

Wherein, described user is the mobile app that uses _mThe score value that distributes

For $s_{{\mathrm{app}}_m}=s_1\×{\mathrm{read}}_{{\mathrm{app}}_m}+s_2\×{\mathrm{download}}_{{\mathrm{app}}_m}+s_3\×{\mathrm{usetime}}_{{\mathrm{app}}_m};$

Wherein, s ₁Equal 1, s ₂Equal 2, s ₃Equal 1; When the user checks the mobile app of application _mThe time

Equal 1, when the user does not check the mobile app of application _mThe time

Equal 0; When the user downloads the mobile app of application _mThe time

Equal 1, when the user does not download the mobile app of application _mThe time

Equal 0;

Equal the user and use the mobile app of application _mDuration.

Wherein, the degree of correlation between the mobile applicating category of described the second computing unit 23 calculating specifically comprises:

Utilize following formula to calculate mobile applicating category concept _iWith mobile applicating category concept _jBetween the degree of correlation:

$R({\mathrm{concept}}_i,{\mathrm{concept}}_j)=\underset{{\mathrm{concept}}_{{\mathrm{app}}_n}={\mathrm{concept}}_j}{\overset{U}{\underset{{\mathrm{concept}}_{{\mathrm{app}}_m}={\mathrm{concept}}_i,}{\mathrm{\Σ}}}}\frac{f_{{\mathrm{app}}_m{\mathrm{app}}_n}}{f_{{\mathrm{app}}_m}+f_{{\mathrm{app}}_n}}\×R({\mathrm{app}}_m,{\mathrm{app}}_n)$

Wherein, concept _iAnd concept _jBe respectively the mobile app of application _mWith the mobile app that uses _nAffiliated mobile applicating category, R (app _m, app _n) be the mobile app of application in the mobile set of applications _mWith the mobile app that uses _nBetween the degree of correlation,

The mobile app of application is checked in expression _mThe user, download the mobile app of application _mThe user and use the mobile app of application _mUser's sum,

The mobile app of application is checked in expression _nThe user, download the mobile app of application _nThe user and use the mobile app of application _nUser's sum,

The mobile app of application is checked in expression _mThe user, download the mobile app of application _mThe user and use the mobile app of application _mThe user set with check the mobile app of application _nThe user, download the mobile app of application _nThe user or use the mobile app of application _nUser's intersection of sets total number of users that collection comprises.

This system also comprises for the 4th computing unit 25 that generates in advance the mobile weighted value of using;

Described the 4th computing unit 25 generates in advance the mobile weighted value of using and specifically comprises:

Utilize following formula to calculate the mobile app of application _mAt mobile applicating category concept _iIn weighted value:

$w_{{\mathrm{concept}}_i{\mathrm{app}}_m}=k_1\×\frac{r_{{\mathrm{app}}_m}}{r_{{\mathrm{concept}}_i}}+k_2\×\frac{d_{{\mathrm{app}}_m}}{d_{{\mathrm{concept}}_i}}+k_3\×\frac{u_{{\mathrm{app}}_m}}{u_{{\mathrm{concept}}_i}}---\left(5\right)$

Wherein,

Be respectively the mobile app of application _mIn user's history log by the total degree checked, the total degree that is downloaded, total duration of being used;

Be respectively mobile applicating category concept _iTotal duration that lower all movements are applied in the total degree of being checked in user's history log, the total degree that is downloaded, are used; k ₁Equal 0.2, k ₂Equal 0.4, k ₃Equal 0.4.

This system also comprises: updating block 26;

Described updating block 26 adds the ontology library of mobile application to for the mobile application that mobile application shop is increased newly, and is that classification information and the attribute information that marks correspondence used in newly-increased moving;

Described the 4th computing unit 25, also on duty with default decay factor with the forward mobile average weight of using of rank under the affiliated mobile applicating category of newly-increased mobile application, the weighted value that moves application that obtains increasing newly.

The recommendation degree that described the first computing unit 21 calculates mobile application under the described mobile applicating category specifically comprises:

Utilize following formula to calculate that each moves the recommendation degree of application in the mobile applicating category:

${\mathrm{rec}}_{{\mathrm{app}}_m{\mathrm{app}}_n}=R({\mathrm{concept}}_i,{\mathrm{concept}}_j)\×w_{{\mathrm{concept}}_j{\mathrm{app}}_n}+k\×\mathrm{comatt}({\mathrm{app}}_m,{\mathrm{app}}_n)$

Wherein,

For specifying the mobile app of application _mThe time recommend the mobile app of application to the user _nThe recommendation degree, the mobile app that uses _mAffiliated mobile applicating category is concept _i, the mobile app that uses _nAffiliated mobile applicating category is concept _j, mobile applicating category concept _jBe positioned at and mobile applicating category concept _iIn the highest mobile applicating category of the degree of correlation, R (concept _i, concept _j) be mobile applicating category concept _iWith mobile applicating category concept _jBetween the degree of correlation, Be mobile applicating category concept _jThe lower mobile app that uses _nWeighted value, comatt (app _m, app _n) use app for moving _nWith the mobile app that uses _mThe number of same alike result, k equals 2.

Technique scheme of the present invention has following useful technique effect:

1, extracts forward mobile application of recommendation degree rank under the higher mobile applicating category of the degree of correlation, and mobile application that wherein degree of recommendation is the highest recommended the user, so, guaranteed the diversity of the mobile applicating category recommended, thus the mobile diversity of using of can Effective Raise recommending.

2, after the calculating of the degree of correlation of mobile applicating category is complete, add mobile application that increases newly to ontology library, and configuration categories information and attribute information, therefore when the recommendation of calculating mobile application is spent, just can include mobile application that increases newly in computer capacity according to ontology library, effectively calculate the newly-increased mobile recommendation degree of using, the mobile application that also can effectively will increase newly according to the recommendation degree is pushed to the user, thereby can effectively solve the newly-increased mobile cold start-up problem of using.

The above only is preferred embodiment of the present invention, and is in order to limit the present invention, within the spirit and principles in the present invention not all, any modification of making, is equal to replacement, improvement etc., all should be included within the scope of protection of the invention.

Claims (18)

1. method for pushing that moves application is characterized in that the method comprises:

According to the degree of correlation of the mobile applicating category that generates in advance, determine one or more the mobile applicating category the highest with the mobile mobile applicating category degree of correlation of using of user's appointment;

According to the mobile weighted value of using that generates in advance, calculate the mobile recommendation degree of using under the described mobile applicating category;

Extract each and move forward mobile application of recommendation degree rank under the applicating category, the mobile application more than that the recommendation degree is the highest in mobile application that the default recommendation results number of foundation will be extracted is pushed to the user as recommendation results.

2. method according to claim 1 is characterized in that, the method that generates in advance the degree of correlation of mobile applicating category is:

Obtain the mobile classification information of using according to the mobile ontology library of using, according to the mobile classification information of using mobile application that the user checks, downloads and uses classified;

According to the user that obtains mobile use check or download in the shop mobile information, the user who uses uses the mobile duration information of using and mobile application that generate in advance between the degree of correlation, calculate the degree of correlation between the mobile applicating category.

3. method according to claim 2 is characterized in that, the method that generates in advance the degree of correlation between mobile the application is:

In mobile application shop, check or download the information and the user that move application according to described user and use the mobile duration information of using, and move the degree of correlation between the application in the mobile set of applications of utilizing following formula calculating user to check, download and use:

$R({\mathrm{app}}_m,{\mathrm{app}}_n)=\underset{u=1}{\overset{U}{\mathrm{\Σ}}}{w}_u\×\frac{s_{{\mathrm{app}}_m}\×(k_1+1)}{s_{{\mathrm{app}}_m}+K}\×\frac{s_{{\mathrm{app}}_n}\×(k_2+1)}{s_{{\mathrm{app}}_n}+k_2}$

Wherein, R (app _m, app _n) represent to move in the mobile set of applications and use app _mWith the mobile app that uses _nBetween the degree of correlation, U represents to use simultaneously the mobile app of application _mWith the mobile app that uses _nUser set, With

Represent that respectively the user gathers that user u is app among the U _mAnd app _nThe score value that distributes; w _uThe expression user gathers the weight of user u among the U,

k ₁Equal 2, k ₂Equal 1.2, b and equal 0.75, n _uThe expression user gathers the mobile sum of using that user u checks, downloads and uses among the U, n _AvgThe mobile sum of using of the mobile sum of using that expression user u checks, download and the mean value that uses the mobile sum of using.

4. method according to claim 3 is characterized in that, described user gathers the weight w of user u among the U _uFor Wherein, N represents the mobile sum of using in the mobile set of applications, n _uThe expression user gathers the mobile sum of using that user u checks, downloads and uses among the U.

5. method according to claim 3 is characterized in that, described user is the mobile app that uses _mThe score value that distributes

For $s_{{\mathrm{app}}_m}=s_1\×{\mathrm{read}}_{{\mathrm{app}}_m}+s_2\×{\mathrm{download}}_{{\mathrm{app}}_m}+s_3\×{\mathrm{usetime}}_{{\mathrm{app}}_m};$

Wherein, s ₁Equal 1, s ₂Equal 2, s ₃Equal 1; When the user checks the mobile app of application _mThe time

Equal 1, when the user does not check the mobile app of application _mThe time

Equal 0; When the user downloads the mobile app of application _mThe time

Equal 1, when the user does not download the mobile app of application _mThe time

Equal 0; Equal the user and use the mobile app of application _mDuration.

6. method according to claim 2 is characterized in that, the method for the degree of correlation between the mobile applicating category of described calculating is:

Utilize following formula to calculate mobile applicating category concept _iWith mobile applicating category concept _jBetween the degree of correlation:

$R({\mathrm{concept}}_i,{\mathrm{concept}}_j)=\underset{{\mathrm{concept}}_{{\mathrm{app}}_n}={\mathrm{concept}}_j}{\overset{U}{\underset{{\mathrm{concept}}_{{\mathrm{app}}_m}={\mathrm{concept}}_i,}{\mathrm{\Σ}}}}\frac{f_{{\mathrm{app}}_m{\mathrm{app}}_n}}{f_{{\mathrm{app}}_m}+f_{{\mathrm{app}}_n}}\×R({\mathrm{app}}_m,{\mathrm{app}}_n)$

Wherein, concept _iAnd concept _jBe respectively the mobile app of application _mWith the mobile app that uses _nAffiliated mobile applicating category, R (app _m, app _n) be the mobile app of application in the mobile set of applications _mWith the mobile app that uses _nBetween the degree of correlation,

The mobile app of application is checked in expression _mThe user, download the mobile app of application _mThe user and use the mobile app of application _mUser's sum,

The mobile app of application is checked in expression _nThe user, download the mobile app of application _nThe user and use the mobile app of application _nUser's sum,

The mobile app of application is checked in expression _mThe user, download the mobile app of application _mThe user and use the mobile app of application _mThe user set with check the mobile app of application _nThe user, download the mobile app of application _nThe user or use the mobile app of application _nUser's intersection of sets total number of users that collection comprises.

7. method according to claim 1 is characterized in that, the method that generates in advance the mobile weighted value of using is:

Utilize following formula to calculate the mobile app of application _mAt mobile applicating category concept _iIn weighted value:

$w_{{\mathrm{concept}}_i{\mathrm{app}}_m}=k_1\×\frac{r_{{\mathrm{app}}_m}}{r_{{\mathrm{concept}}_i}}+k_2\×\frac{d_{{\mathrm{app}}_m}}{d_{{\mathrm{concept}}_i}}+k_3\×\frac{u_{{\mathrm{app}}_m}}{u_{{\mathrm{concept}}_i}}$

Wherein,

Be respectively the mobile app of application _mIn user's history log by the total degree checked, the total degree that is downloaded, total duration of being used;

Be respectively mobile applicating category concept _iTotal duration that lower all movements are applied in the total degree of being checked in user's history log, the total degree that is downloaded, are used; k ₁Equal 0.2, k ₂Equal 0.4, k ₃Equal 0.4.

8. method according to claim 1 is characterized in that, the method also comprises:

With the mobile ontology library that adds mobile application to of using that increases newly in the mobile application shop, and be newly-increased mobile classification information and the attribute information that marks correspondence of using;

On duty with default decay factor with the mobile average weight of using that rank under the mobile applicating category under newly-increased mobile the application is forward, the mobile weighted value of using that obtains increasing newly.

9. method according to claim 1 is characterized in that, the method for the mobile recommendation degree of using is under the described mobile applicating category of described calculating:

Utilize following formula to calculate that each moves the recommendation degree of application in the mobile applicating category:

${\mathrm{rec}}_{{\mathrm{app}}_m{\mathrm{app}}_n}=R({\mathrm{concept}}_i,{\mathrm{concept}}_j)\×w_{{\mathrm{concept}}_j{\mathrm{app}}_n}+k\×\mathrm{comatt}({\mathrm{app}}_m,{\mathrm{app}}_n)$

Wherein,

For specifying the mobile app of application _mThe time recommend the mobile app of application to the user _nThe recommendation degree, the mobile app that uses _mAffiliated mobile applicating category is concept _i, the mobile app that uses _nAffiliated mobile applicating category is concept _j, mobile applicating category concept _jBe positioned at and mobile applicating category concept _iIn the highest mobile applicating category of the degree of correlation, R (concept _i, concept _j) be mobile applicating category concept _iWith mobile applicating category concept _jBetween the degree of correlation,

Be mobile applicating category concept _jThe lower mobile app that uses _nWeighted value, comatt (app _m, app _n) use app for moving _nWith the mobile app that uses _mThe number of same alike result, k equals 2.

10. a supplying system that moves application is characterized in that, this system comprises: statistic unit, the first computing unit, push unit; Wherein,

Statistic unit is used for the degree of correlation according to the mobile applicating category that generates in advance, determines one or more the mobile applicating category the highest with the mobile mobile applicating category degree of correlation of using of user's appointment;

The first computing unit is used for calculating the mobile recommendation degree of using under the described mobile applicating category according to the mobile weighted value of using that generates in advance;

Push unit is used for extracting each and moves forward mobile application of recommendation degree rank under the applicating category, and the mobile application more than that the recommendation degree is the highest in mobile application that the default recommendation results number of foundation will be extracted is pushed to the user as recommendation results.

11. system according to claim 10 is characterized in that, this system also comprises: the second computing unit that is used for generating in advance the degree of correlation of mobile applicating category:

The degree of correlation that the second computing unit generates mobile applicating category in advance specifically comprises: obtain the mobile classification information of using according to the mobile ontology library of using, according to the mobile classification information of using mobile application that the user checks, downloads and uses classified; According to the user that obtains mobile use check or download in the shop mobile information, the user who uses uses the mobile duration information of using and mobile application that generate in advance between the degree of correlation, calculate the degree of correlation between the mobile applicating category.

12. system according to claim 11 is characterized in that, this system also comprises for the 3rd computing unit that generates in advance the degree of correlation between mobile the application;

The degree of correlation that described the 3rd computing unit generates between mobile the application in advance specifically comprises:

In mobile application shop, check or download the information and the user that move application according to described user and use the mobile duration information of using, and move the degree of correlation between the application in the mobile set of applications of utilizing following formula calculating user to check, download and use:

$R({\mathrm{app}}_m,{\mathrm{app}}_n)=\underset{u=1}{\overset{U}{\mathrm{\Σ}}}{w}_u\×\frac{s_{{\mathrm{app}}_m}\×(k_1+1)}{s_{{\mathrm{app}}_m}+K}\×\frac{s_{{\mathrm{app}}_n}\×(k_2+1)}{s_{{\mathrm{app}}_n}+k_2}$

Wherein, R (app _m, app _n) represent to move in the mobile set of applications and use app _mWith the mobile app that uses _nBetween the degree of correlation, U represents to use simultaneously the mobile app of application _mWith the mobile app that uses _nUser set,

With

Represent that respectively the user gathers that user u is app among the U _mAnd app _nThe score value that distributes; w _uThe expression user gathers the weight of user u among the U,

k ₁Equal 2, k ₂Equal 1.2, b and equal 0.75, n _uThe expression user gathers the mobile sum of using that user u checks, downloads and uses among the U, n _AvgThe mobile sum of using of the mobile sum of using that expression user u checks, download and the mean value that uses the mobile sum of using.

13. system according to claim 12 is characterized in that, described user gathers the weight w of user u among the U _uFor

Wherein, N represents the mobile sum of using in the mobile set of applications, n _uThe expression user gathers the mobile sum of using that user u checks, downloads and uses among the U.

14. system according to claim 12 is characterized in that, described user is the mobile app that uses _mThe score value that distributes

For $s_{{\mathrm{app}}_m}=s_1\×{\mathrm{read}}_{{\mathrm{app}}_m}+s_2\×{\mathrm{download}}_{{\mathrm{app}}_m}+s_3\×{\mathrm{usetime}}_{{\mathrm{app}}_m};$

Wherein, s ₁Equal 1, s ₂Equal 2, s ₃Equal 1; When the user checks the mobile app of application _mThe time

Equal 1, when the user does not check the mobile app of application _mThe time Equal 0; When the user downloads the mobile app of application _mThe time

Equal 1, when the user does not download the mobile app of application _mThe time Equal 0;

Equal the user and use the mobile app of application _mDuration.

15. system according to claim 11 is characterized in that, the degree of correlation that described the second computing unit calculates between the mobile applicating category specifically comprises:

Utilize following formula to calculate mobile applicating category concept _iWith mobile applicating category concept _jBetween the degree of correlation:

$R({\mathrm{concept}}_i,{\mathrm{concept}}_j)=\underset{{\mathrm{concept}}_{{\mathrm{app}}_n}={\mathrm{concept}}_j}{\overset{U}{\underset{{\mathrm{concept}}_{{\mathrm{app}}_m}={\mathrm{concept}}_i,}{\mathrm{\Σ}}}}\frac{f_{{\mathrm{app}}_m{\mathrm{app}}_n}}{f_{{\mathrm{app}}_m}+f_{{\mathrm{app}}_n}}\×R({\mathrm{app}}_m,{\mathrm{app}}_n)$

Wherein, concept _iAnd concept _jBe respectively the mobile app of application _mWith the mobile app that uses _nAffiliated mobile applicating category, R (app _m, app _n) be the mobile app of application in the mobile set of applications _mWith the mobile app that uses _nBetween the degree of correlation,

The mobile app of application is checked in expression _mThe user, download the mobile app of application _mThe user and use the mobile app of application _mUser's sum,

The mobile app of application is checked in expression _nThe user, download the mobile app of application _nThe user and use the mobile app of application _nUser's sum,

The mobile app of application is checked in expression _mThe user, download the mobile app of application _mThe user and use the mobile app of application _mThe user set with check the mobile app of application _nThe user, download the mobile app of application _nThe user or use the mobile app of application _nUser's intersection of sets total number of users that collection comprises.

16. system according to claim 10 is characterized in that, this system also comprises for the 4th computing unit that generates in advance the mobile weighted value of using;

Described the 4th computing unit generates in advance the mobile weighted value of using and specifically comprises:

Utilize following formula to calculate the mobile app of application _mAt mobile applicating category concept _iIn weighted value:

$w_{{\mathrm{concept}}_i{\mathrm{app}}_m}=k_1\×\frac{r_{{\mathrm{app}}_m}}{r_{{\mathrm{concept}}_i}}+k_2\×\frac{d_{{\mathrm{app}}_m}}{d_{{\mathrm{concept}}_i}}+k_3\×\frac{u_{{\mathrm{app}}_m}}{u_{{\mathrm{concept}}_i}}$

Wherein,

Be respectively the mobile app of application _mIn user's history log by the total degree checked, the total degree that is downloaded, total duration of being used;

Be respectively mobile applicating category concept _iTotal duration that lower all movements are applied in the total degree of being checked in user's history log, the total degree that is downloaded, are used; k ₁Equal 0.2, k ₂Equal 0.4, k ₃Equal 0.4.

17. system according to claim 10 is characterized in that, this system also comprises: updating block;

Described updating block adds the ontology library of mobile application to for the mobile application that mobile application shop is increased newly, and is that classification information and the attribute information that marks correspondence used in newly-increased moving;

Described the 4th computing unit, also on duty with default decay factor with the forward mobile average weight of using of rank under the affiliated mobile applicating category of newly-increased mobile application, the weighted value that moves application that obtains increasing newly.

18. system according to claim 10 is characterized in that, the recommendation degree that described the first computing unit calculates mobile application under the described mobile applicating category specifically comprises:

Utilize following formula to calculate that each moves the recommendation degree of application in the mobile applicating category:

${\mathrm{rec}}_{{\mathrm{app}}_m{\mathrm{app}}_n}=R({\mathrm{concept}}_i,{\mathrm{concept}}_j)\×w_{{\mathrm{concept}}_j{\mathrm{app}}_n}+k\×\mathrm{comatt}({\mathrm{app}}_m,{\mathrm{app}}_n)$

Wherein,

For specifying the mobile app of application _mThe time recommend the mobile app of application to the user _nThe recommendation degree, the mobile app that uses _mAffiliated mobile applicating category is concept _i, the mobile app that uses _nAffiliated mobile applicating category is concept _j, mobile applicating category concept _jBe positioned at and mobile applicating category concept _iIn the highest mobile applicating category of the degree of correlation, R (concept _i, concept _j) be mobile applicating category concept _iWith mobile applicating category concept _jBetween the degree of correlation,

Be mobile applicating category concept _jThe lower mobile app that uses _nWeighted value, comatt (app _m, app _n) use app for moving _nWith the mobile app that uses _mThe number of same alike result, k equals 2.

Images